PROJECT SUMMARY/ABSTRACT This SHINE II proposal seeks to promote broader research into the role of RNA processing in erythropoiesis. The broad hypothesis of this work is that a highly orchestrated RNA processing program is essential for normal erythropoiesis, with the critical corollary that defects in RNA processing should have major adverse impacts on differentiation and/or red cell function that may underlie unexplained erythroid disorders. These concepts are actively exploited to provide exciting new insights in nonerythroid biology and disease, but until the recent discovery of splicing machinery mutations in myelodysplasia and other hematologic cancers, RNA processing has been under-appreciated in erythropoiesis. To study the impact of RNA processing in an erythroid in vivo system, this proposal focuses on a knockout mouse model with complete deficiency of MBNL1 (Muscleblind- like1), a zinc finger protein that regulates post-transcriptional processes including alternative pre-mRNA splicing (via binding to intron regulatory elements) and mRNA stability (via binding to 3'UTR sequences). Functional depletion of MBNL1 in the triplet repeat disease, myotonic dystrophy, causes splicing defects that correlate with specific physiological deficits of the disease. Preliminary data show that differentiating erythroblasts execute a robust alternative splicing program; that MBNL1 is one of the most abundant RNA processing factors in normal erythroblasts of both human and mouse; that MBNL1 can regulate a key alternative splicing event (in protein 4.1R pre-mRNA) required for mechanically stable red cell membranes; and that MBNL1 knockout mice have erythroid deficits manifested by elevated reticulocytes, reduced hematocrit, and greatly enlarged spleen. These observations support the hypothesis that MBNL1 deficiency should have a major impact on the erythroblast transcriptome, as it does in muscle. In accordance with the SHINE II format, this proposal consists of one aim. Aim 1A will be a global transcriptome analysis of FACS-purified proerythro- blasts, orthochromatic erythroblasts, and reticulocytes using state of the art RNA-seq and computational analytical strategies. Comparison of MBNL1-knockouts with normal littermates is expected to reveal dozens to hundreds of altered transcripts that will represent candidate mediators of disease in these animals. Aim 1B will validate MBNL1-mediated RNA processing events predicted in 1A, using minigene approaches that allow manipulation of MBNL1 expression and MBNL1 binding motifs. Expected outcomes of this research are novel insights into MBNL1-regulated RNA processing networks, and identification of erythroid processes impacted by these networks that may contribute to pathology in knockout animals. Besides MBNL1, several other multi- functional RBPs with post-transcriptional regulatory activities and documented disease involvement in nonerythroid tissues are abundantly expressed in erythroid cells. Defects in this class of proteins may be an unappreciated cause of erythroid disease. Understanding splicing events critical for erythroid function could facilitate future therapeutic intervention with antisense technologies being developed to treat splicing diseases.